There is currently no cure for retinitis pigmentosa (RP), which is the most prevalent inherited retinal disease (IRD). RP patietns first suffer from nightblindness caused by expression of disease genes in rods, then gradually lose their peripheral daylight vision, mediated by cones, thus forming tunnel vision. They eventually become legally blind with the complete loss of central cone-mediated vision. It is unknown why the cones lose their function and eventually die after rods die. Rescuing cones from the non-autonomous degeneration has been proposed as a generic therapeutic approach to treat RP, regardless of the rod gene mutation. Recently, the applicant found an effective therapeutic gene, Txnip, out of 30 candidates using AAV-mediated gene therapy in RP mouse models. Txnip is excellent in delaying RP cone degeneration after the rods are gone. However, it is unclear why Txnip can rescue the RP cones from degeneration. Therefore, the applicant propose to study the mechanism of Txnip in the mentored phase (K99), during which the applicant will continue his training in molecular biology, and in two-photon adaptive optics scanning light ophthalmoloscopy (2P-AOSLO) that will allow dissecting the downstream targets of this therapeutic candidate. Furthermore, in the independent phase (R00), the applicant propose to investigate in the mechanism of cone functional loss, which occurs abruptly upon rod death, but well before cone degeneration. Specifically, the applicant will apply shRNA knockdown of proposed downstream genes of the rescue gene, co-deliver the knockdown with the rescue target, and examine if the rescue effect on RP cone survival can be abolished (Aim I, K99). The applicant will also examine the proposed metabolite level changes in RP cones with and without this therapeutic gene by imaging of metabolite sensors ex vivo and in vivo (Aim II, K99). Finally, the applicant will investigate the mechanism of the rapid RP cone ERG function loss based on preliminary electrophysiology, RNA-seq and histology data, in the hope of preserving the RP cone function as long as possible before their eventual death (Aim III, R00). Overall, this project is designed to combine molecular biology, high-resolution imaging, and electrophysiology to find a treatment for RP. The mentored phase of this award will be conducted under the supervision of Dr. Constance Cepko (Harvard), and the co-mentorship of Dr. Jennifer Hunter (U Rochester) who is an expert in 2P-AOSLO. This project will also receive material and advisory support from Dr. Gary Yellen (Harvard), who is an expert in genetically encoded metabolite sensors.